Age-related macular degeneration (AMD) is the leading cause of irreversible blindness in the western world affecting nearly 30% of those over the age of 75. The only accepted risk factors for AMD are age, race, and smoking. AMD alters the quality of life of those affected by causing a debilitating loss of central vision. Currently diagnosis of AMD at its early stages can be difficult. This has led to problems in the development of drugs to treat the disease since clinical trials must rely on following the slow progression of the disease in individuals that are already demonstrating some loss of visual acuity. We have recently found that Bruch's membrane and sub-retinal pigment epithelium (RPE) deposits characteristic of AMD exhibit a unique autofluorescence spectrum that can be excited with light between 360 and 490nm. When compared with the intensity of autofluorescence elicited from RPE lipofuscin, these emissions were significantly elevated in eyes from donors with AMD versus healthy donor eyes. The change in the ratios of these emissions (Bruch's membrane and sub-RPE deposits vs. lipofuscin) correlating with disease suggests that the same principles that allow quantitative microscopy using fluorescence ratiometry could be applied to fundus photography. The goal of the work outlined in this proposal is to design, build, and test a device for fundus fluorescence ratiometry, and to determine the efficacy of this device in the early diagnosis of retinal lesions that are risk factors for AMD. To accomplish this we will perform a series of 3 specific aims. In the first aim we will modify a standard clinical fundus camera for fundus fluorescence ratiometry. In specific aim 2 we will determine ideal excitation and emission wavelengths for fundus fluorescence ratiometry. This will be accomplished by testing a variety of excitation and emission wavelengths using the modified camera on postmortem human donor eyes. In specific aim 3 we will determine whether fundus fluorescence ratiometry shows promise for the diagnosis of lesions associated with AMD. This will be accomplished using eyes from donors with AMD. Specifically we will test the efficacy of this device in identifying basal deposits. Basal deposits are specific retinal lesions that are not visible during a clinical fundus exam. One form of basal deposit, basal linear deposits, are more common in AMD eyes than age-matched controls. A positive outcome to this study would warrant future clinical trials.